FEATURE: Replace SimpleRSS with Ruby RSS module (#5311)

* SPEC: PollFeedJob parsing atom feed

* add FeedItemAccessor

It is to provide a consistent interface to access a feed item's tag
content.

* add FeedElementInstaller

to install non-standard and non-namespaced feed elements

* FEATURE: replace SimpleRSS with Ruby RSS module

* get FinalDestination and download with Excon

* support namespaced element with FeedElementInstaller

Gemfile

@@ -166,7 +166,6 @@ gem 'gc_tracer', require: false, platform: :mri
 
 # required for feed importing and embedding
 gem 'ruby-readability', require: false
-gem 'simple-rss', require: false
 
 gem 'stackprof', require: false, platform: :mri
 gem 'memory_profiler', require: false, platform: :mri

Gemfile.lock

@@ -368,7 +368,6 @@ GEM
       connection_pool (~> 2.2, >= 2.2.0)
       rack-protection (>= 1.5.0)
       redis (>= 3.3.4, < 5)
-    simple-rss (1.3.1)
     slop (3.6.0)
     sprockets (3.7.1)
       concurrent-ruby (~> 1.0)
@@ -500,7 +499,6 @@ DEPENDENCIES
   seed-fu
   shoulda
   sidekiq
-  simple-rss
   sprockets-rails
   stackprof
   thor

app/jobs/scheduled/poll_feed.rb

@@ -2,9 +2,13 @@
 # Creates and Updates Topics based on an RSS or ATOM feed.
 #
 require 'digest/sha1'
+require 'excon'
+require 'rss'
+require_dependency 'feed_item_accessor'
+require_dependency 'feed_element_installer'
+require_dependency 'final_destination'
 require_dependency 'post_creator'
 require_dependency 'post_revisor'
-require 'open-uri'
 
 module Jobs
   class PollFeed < Jobs::Scheduled
@@ -46,17 +50,11 @@ def import_topics(feed_topics)
 
     def import_topic(topic)
       if topic.user
-        TopicEmbed.import(topic.user, topic.url, topic.title, CGI.unescapeHTML(topic.content.scrub))
+        TopicEmbed.import(topic.user, topic.url, topic.title, CGI.unescapeHTML(topic.content))
       end
     end
 
     class Feed
-      require 'simple-rss'
-
-      if SiteSetting.embed_username_key_from_feed.present?
-        SimpleRSS.item_tags << SiteSetting.embed_username_key_from_feed.to_sym
-      end
-
       def initialize
         @feed_url = SiteSetting.feed_polling_url
         @feed_url = "http://#{@feed_url}" if @feed_url !~ /^https?\:\/\//
@@ -65,7 +63,7 @@ def initialize
       def topics
         feed_topics = []
 
-        rss = fetch_rss
+        rss = parsed_feed
         return feed_topics unless rss.present?
 
         rss.items.each do |i|
@@ -78,36 +76,56 @@ def topics
 
       private
 
-      def fetch_rss
-        SimpleRSS.parse open(@feed_url, allow_redirections: :all)
-      rescue OpenURI::HTTPError, SimpleRSSError
+      def parsed_feed
+        raw_feed = fetch_rss
+        return nil if raw_feed.blank?
+
+        if SiteSetting.embed_username_key_from_feed.present?
+          FeedElementInstaller.install(SiteSetting.embed_username_key_from_feed, raw_feed)
+        end
+
+        RSS::Parser.parse(raw_feed)
+      rescue RSS::NotWellFormedError, RSS::InvalidRSSError
         nil
       end
 
+      def fetch_rss
+        final_destination = FinalDestination.new(@feed_url, verbose: true)
+        feed_final_url = final_destination.resolve
+        return nil unless final_destination.status == :resolved
+
+        Excon.new(feed_final_url.to_s).request(method: :get, expects: 200).body
+      rescue Excon::Error::HTTPStatus
+        nil
+      end
     end
 
     class FeedTopic
       def initialize(article_rss_item)
-        @article_rss_item = article_rss_item
+        @accessor = FeedItemAccessor.new(article_rss_item)
       end
 
       def url
-        link = @article_rss_item.link
+        link = @accessor.link
         if url?(link)
           return link
         else
-          return @article_rss_item.id
+          return @accessor.element_content(:id)
         end
       end
 
       def content
-        @article_rss_item.content_encoded&.force_encoding("UTF-8")&.scrub ||
-          @article_rss_item.content&.force_encoding("UTF-8")&.scrub ||
-          @article_rss_item.description&.force_encoding("UTF-8")&.scrub
+        content = nil
+
+        %i[content_encoded content description].each do |content_element_name|
+          content ||= @accessor.element_content(content_element_name)
+        end
+
+        content&.force_encoding('UTF-8')&.scrub
       end
 
       def title
-        @article_rss_item.title.force_encoding("UTF-8").scrub
+        @accessor.element_content(:title).force_encoding('UTF-8').scrub
       end
 
       def user
@@ -125,11 +143,7 @@ def url?(link)
       end
 
       def author_username
-        begin
-          @article_rss_item.send(SiteSetting.embed_username_key_from_feed.to_sym)
-        rescue
-          nil
-        end
+        @accessor.element_content(SiteSetting.embed_username_key_from_feed.sub(':', '_'))
       end
 
       def default_user
@@ -145,9 +159,6 @@ def author_user
       def find_user(user_name)
         User.where(username_lower: user_name).first
       end
-
     end
-
   end
-
 end

lib/feed_element_installer.rb

@@ -0,0 +1,52 @@
+require 'rexml/document'
+require 'rss'
+
+class FeedElementInstaller
+  private_class_method :new
+
+  def self.install(element_name, feed)
+    # RSS Specification at http://cyber.harvard.edu/rss/rss.html#extendingRss
+    # > A RSS feed may contain [non-standard elements], only if those elements are *defined in a namespace*
+
+    new(element_name, feed).install if element_name.include?(':')
+  end
+
+  attr_reader :feed, :original_name, :element_namespace, :element_name, :element_accessor
+
+  def initialize(element_name, feed)
+    @feed = feed
+    @original_name = element_name
+    @element_namespace, @element_name = *element_name.split(':')
+    @element_accessor = "#{@element_namespace}_#{@element_name}"
+  end
+
+  def element_uri
+    @element_uri ||= REXML::Document.new(feed).root&.attributes&.namespaces&.fetch(@element_namespace, '') || ''
+  end
+
+  def install
+    install_in_rss unless installed_in_rss?
+    install_in_atom unless installed_in_atom?
+  end
+
+  private
+
+  def install_in_rss
+    RSS::Rss::Channel::Item.install_text_element(element_name, element_uri, '?', element_accessor, nil, original_name)
+    RSS::BaseListener.install_get_text_element(element_uri, element_name, element_accessor)
+  end
+
+  def install_in_atom
+    RSS::Atom::Entry.install_text_element(element_name, element_uri, '?', element_accessor, nil, original_name)
+    RSS::Atom::Feed::Entry.install_text_element(element_name, element_uri, '?', element_accessor, nil, original_name)
+    RSS::BaseListener.install_get_text_element(element_uri, element_name, element_accessor)
+  end
+
+  def installed_in_rss?
+    RSS::Rss::Channel::Item.method_defined?(element_accessor)
+  end
+
+  def installed_in_atom?
+    RSS::Atom::Entry.method_defined?(element_accessor) || RSS::Atom::Feed::Entry.method_defined?(element_accessor)
+  end
+end

lib/feed_item_accessor.rb

@@ -0,0 +1,25 @@
+class FeedItemAccessor
+  attr_accessor :rss_item
+
+  def initialize(rss_item)
+    @rss_item = rss_item
+  end
+
+  def element_content(element_name)
+    try_attribute_or_self(element(element_name), :content)
+  end
+
+  def link
+    try_attribute_or_self(element(:link), :href)
+  end
+
+  private
+
+  def element(element_name)
+    rss_item.respond_to?(element_name) ? rss_item.send(element_name) : nil
+  end
+
+  def try_attribute_or_self(element, attribute_name)
+    element.respond_to?(attribute_name) ? element.send(attribute_name) : element
+  end
+end

spec/components/feed_element_installer_spec.rb

@@ -0,0 +1,40 @@
+require 'feed_element_installer'
+require 'rails_helper'
+
+describe FeedElementInstaller do
+  describe '#install_rss_element' do
+    let(:raw_feed) { file_from_fixtures('feed.rss', 'feed').read }
+
+    it 'creates parsing for a non-standard, namespaced element' do
+      FeedElementInstaller.install('discourse:username', raw_feed)
+      feed = RSS::Parser.parse(raw_feed)
+
+      expect(feed.items.first.discourse_username).to eq('xrav3nz')
+    end
+
+    it 'does not create parsing for a non-standard, non-namespaced element' do
+      FeedElementInstaller.install('username', raw_feed)
+      feed = RSS::Parser.parse(raw_feed)
+
+      expect { feed.items.first.username }.to raise_error(NoMethodError)
+    end
+  end
+
+  describe '#install_atom_element' do
+    let(:raw_feed) { file_from_fixtures('feed.atom', 'feed').read }
+
+    it 'creates parsing for a non-standard, namespaced element' do
+      FeedElementInstaller.install('discourse:username', raw_feed)
+      feed = RSS::Parser.parse(raw_feed)
+
+      expect(feed.items.first.discourse_username).to eq('xrav3nz')
+    end
+
+    it 'does not create parsing for a non-standard, non-namespaced element' do
+      FeedElementInstaller.install('username', raw_feed)
+      feed = RSS::Parser.parse(raw_feed)
+
+      expect { feed.items.first.username }.to raise_error(NoMethodError)
+    end
+  end
+end

spec/components/feed_item_accessor_spec.rb

@@ -0,0 +1,33 @@
+require 'rss'
+require 'feed_item_accessor'
+require 'rails_helper'
+
+describe FeedItemAccessor do
+  context 'for ATOM feed' do
+    let(:atom_feed) { RSS::Parser.parse(file_from_fixtures('feed.atom', 'feed'), false) }
+    let(:atom_feed_item) { atom_feed.items.first }
+    let(:item_accessor) { FeedItemAccessor.new(atom_feed_item) }
+
+    describe '#element_content' do
+      it { expect(item_accessor.element_content('title')).to eq(atom_feed_item.title.content) }
+    end
+
+    describe '#link' do
+      it { expect(item_accessor.link).to eq(atom_feed_item.link.href) }
+    end
+  end
+
+  context 'for RSS feed' do
+    let(:rss_feed) { RSS::Parser.parse(file_from_fixtures('feed.rss', 'feed'), false) }
+    let(:rss_feed_item) { rss_feed.items.first }
+    let(:item_accessor) { FeedItemAccessor.new(rss_feed_item) }
+
+    describe '#element_content' do
+      it { expect(item_accessor.element_content('title')).to eq(rss_feed_item.title) }
+    end
+
+    describe '#link' do
+      it { expect(item_accessor.link).to eq(rss_feed_item.link) }
+    end
+  end
+end

spec/fixtures/feed/feed.atom

@@ -0,0 +1,30 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<feed
+  xmlns="http://www.w3.org/2005/Atom"
+  xmlns:thr="http://purl.org/syndication/thread/1.0"
+  xmlns:discourse="http://discourse.org/rss/modules/discourse/"
+  xml:lang="en-US"
+  xml:base="https://blog.discourse.org/wp-atom.php"
+   >
+  <title type="text">Discourse</title>
+  <subtitle type="text">Official blog for the open source Discourse project</subtitle>
+  <updated>2017-10-23T23:45:37Z</updated>
+  <link rel="alternate" type="text/html" href="https://blog.discourse.org" />
+  <id>https://blog.discourse.org/feed/atom/</id>
+  <link rel="self" type="application/atom+xml" href="https://blog.discourse.org/feed/atom/" />
+  <generator uri="https://wordpress.org/" version="4.8.2">WordPress</generator>
+  <entry>
+    <discourse:username><![CDATA[xrav3nz]]></discourse:username>
+    <author>
+      <name>xrav3nz</name>
+    </author>
+    <title type="html"><![CDATA[Poll Feed Spec Fixture]]></title>
+    <link rel="alternate" type="text/html" href="https://blog.discourse.org/2017/09/poll-feed-spec-fixture/" />
+    <id>https://blog.discourse.org/?p=pollfeedspec</id>
+    <updated>2017-09-14T15:22:33Z</updated>
+    <published>2017-09-14T15:22:33Z</published>
+    <category scheme="https://blog.discourse.org" term="design" />
+    <summary type="html"><![CDATA[Here are some random descriptions... [&#8230;]]]></summary>
+    <content type="html" xml:base="https://blog.discourse.org/2017/09/poll-feed-spec-fixture/"><![CDATA[<p>This is the body &amp; content. </p>]]></content>
+  </entry>
+</feed>

spec/fixtures/feed/feed.rss

@@ -5,6 +5,7 @@
   xmlns:atom="http://www.w3.org/2005/Atom"
   xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
   xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
+  xmlns:discourse="http://discourse.org/rss/modules/discourse/"
   >
 <channel>
   <title>Discourse</title>
@@ -21,6 +22,7 @@
     <link>https://blog.discourse.org/2017/09/poll-feed-spec-fixture/</link>
     <pubDate>Thu, 14 Sep 2017 15:22:33 +0000</pubDate>
     <dc:creator><![CDATA[xrav3nz]]></dc:creator>
+    <discourse:username><![CDATA[xrav3nz]]></discourse:username>
     <category><![CDATA[spec]]></category>
     <guid isPermaLink="false">https://blog.discourse.org/?p=pollfeedspec</guid>
     <description><![CDATA[Here are some random descriptions... [&#8230;]]]></description>